The effectiveness of vaccine therapy of cancer is dependent upon the ability to sensitize the host immune system to antigens uniquely expressed by tumor cells. However, previous clinical vaccine trials have been ineffective in demonstrating meaningful antitumor efficacy. The reasons for this ineffectiveness may relate to the weakness of the tumor antigen and/or the presence of tumor-induced immunosuppression. We have demonstrated in animal models that lymph nodes (LN) draining a progressive tumor contain T cells that are immunologically sensitized but functionally deficient in expressing antitumor reactivity. We have designated these cells as "pre-effector" cells since they can be further stimulated in vitro to differentiate into potent effector cells with capacity to mediate tumor regression upon adoptive transfer. The in vitro procedures which promote this differentiation involve culture of pre-effector cells with excessive numbers of tumor cells in the presence of IL-2; or stimulation with anti-CD3 mAb followed by exposure to IL-2. Despite the capability of these culture procedures to selectively expand tumor reactive pre-effector cells, adequate pre-effector cell sensitization in the host is required. For example, poorly immunogenic tumors do not induce pre-effector cells in the tumor-draining LN during progressive tumor growth. However, preeffector cells sensitized to these tumors can be elicited by stimulating the host with tumor cells admixed with a bacterial adjuvant. These findings demonstrate that the pre-effector cell response can be augmented by immunological manipulations. In addition, we found that the pre-effector cell response is subjected to immunologically specific suppression. The ability to retrieve pre-effector cells is diminished as tumors progress and is also subjected to inhibition by the concomitant presence of progressive metastases. We propose to examine the immunological responses induced by tumors modified with genes encoding cytokines and/or allogeneic MHC antigens. Preliminary data suggests that sensitization of T cells can potentially be up-regulated in response to gene-modified tumors, and that tumor-induced suppressive mechanisms can be overcome. Since human tumors are postulated to be poorly immunogenic by virtue of their spontaneous origin, the principles identified in the proposed studies will have clinical application. We plan to evaluate genetically modified tumors as agents for vaccination in order to isolate and propagate antitumor reactive cells for therapy. The specific aims of the proposal are: 1) To elucidate the mechanisms involved in the development of systemic immunity to tumors with diverse immunogenecities by transduction of genes encoded for TNFalpha, IL-2, allogeneic MHC antigen or a combination of these genes, 2) To evaluate the antitumor reactivity of T cells isolated from hosts bearing genetically modified tumors, 3) To evaluate the ability of genetically modified tumors to overcome the suppression of the pre-effector response, and 4) To evaluate the effectiveness of modulating the T cell immune response to tumors utilizing a novel in vivo gene transfer technique.